THE JOURN~ OF BIOLOGICAL CHEMISTRY
Vol. 239, No. 12, December 1964
Printed
in
U.S.A.
Fat
XXIII.
PROPERTIES
Metabolism
OF A SOLUBLE
PETER
From
the Department
of Biochemistry
(Received
FATTY
in Higher
ACID
OVERATHt
SYNTHETASE
AND
P. K.
FROM
AVOCADO
MESOCARP*
STUMPF
and Biophysics,
University
for publication,
June
of California,
Davis,
California
17, 1964)
cation of S-benzoylpantetheine
(mp. 113-116”)2 or by reduction
of n-pantethine
with 2% NaHg in methanol at pH 8 to 9 (23).
Acetyl-CoA,
caproyl-CoA, and caproylpantetheine
were prepared
according to Simon and Shemin (25). The thioesters were
assayed by conversion into the corresponding
hydroxamic acids
(26). Protein was determined
either by the biuret reaction
(27) or spectrophotometrically
(28).
Enzyme Assays
Fatty Acid Synthesis-Typical
conditions for the reaction are
described in the legend to Fig. 2. The incubation was stopped
by the addition of 1.0 ml of 15% methanolic
KOH.
After 3
hours at 70”, the samples were acidified with 6 N HCl and exThe chloroform layer
tracted twice with 10 ml of chloroform.
was washed with 10 ml of a solution containing 1 y0 malonic and
1% acetic acid. An aliquot was then dried in a scintillation
vial and counted in a Packard liquid scintillation
spcctromctcr.
Malonyl-CoA-COz
Exchange-The
assay for the exchange
reaction was performed as described by Alberts, Goldman, and
Vagelos (12). The reaction was stopped by the addition of 0.2
ml of acetic acid. An aliquot was then dried on a strip (2.5 x
7.5 cm) of Whatman No. 1 paper, which was counted directly in
the scintillation
counter.
Enzyme Preparation@
EXPERIMENTAL
Materials
PROCEDURE
and Methods
CoA, glucose g-phosphate, glucose 6-phosphate dehydrogenase,
papain, n-pantethine,
and DPNH were purchased from Sigma
Chemical Company; TPN was purchased from C. F. Boehringer
und Soehne. Ba14C03 was a product of the Oak Ridge National
Laboratories.
Malonyl-2-14C-CoA
was prepared according to
the method of Eggerer and Lynen (23), starting with 100 to 200
mg of malonic acid-2-14C (New England Nuclear Corporation).
Malonyl-CoA
and malonylpantetheine
were prepared by the
same method.
Caproic anhydride was prepared by the method
of Antenriet’h (24). Pantetheine was prepared either by saponifi* This work was supported by grants from the National Science
Foundation
(G-14823) and the National
Institutes
of Health
(GM-10132).
Fellow,
United
States Public
f Foreign Scientist Research
Health
logisches
Service
Institut
(FF
621).
Present
der
Universitaet
address,
Pflanzen
PhysioGoettingen,
Goettingen,
E. coli Enzymes-Fraction
A and enzyme II were prepared
according to Goldman, Alberts, and Vagelos (13). We are
indebted to Dr. Peter Goldman for providing us with a sample
and a culture of E. coli K12 Hfr This- and unpublished
details
for the improved separation and purification
of these enzymes.
The units of activity used are the same as given by Goldman
et al. (13).
Enzyme Preparations from Avocado-Avocados
of the Fuerte
variety were customarily
used, but other varieties were occasionally tested.
For the preparation
of crude extract, 150 g of
mesocarp were homogenized
with 300 ml of 0.02 M potassium
M cysteine, pH 7.3, for 1 minute in a Waring
phosphate-O.001
Blendor.
The homogenate was centrifuged for 10 minutes at
The middle layer of
13,000 x g in a Servall RC-2 centrifuge.
extract was separated from the lipid floating on top and from
the pellet, and centrifuged for 10 minutes at 37,000 X g. Residual lipid and large particles were removed by filtration through
Germany.
Gemeinsame Tagung der Gesellschaft fiir physiologische Chemie und der Oesterreichischen biochemischen Gesellschaft, Vienna, September 26, 1962.
1 F. Lynen,
lecture
given
at the
for
2 We are indebted
to Dr. Alexander
Hagen,
Munich,
sending
us a sample
of S-benzoylpantetheine.
all enzyme preparations
3 Unless
stated
otherwise,
tained at O-4”.
4103
Germany,
were ob-
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
The conversion of malonyl coenzyme A to long chain fatty
acids has been studied in enzyme systems from various tissues
(l-7).
Attempts
to purify the enzymes involved in this conversion revealed an important
difference between these systems.
Thus the highly purified synthetase from yeast (8, 9) and from
liver (10, 11) behaves as a single complex; its fractionation
has
not yet been accomplished.
However, extracts of Clostridium
kluyveri and Escherichia coli can be resolved into several components which on combination
catalyze the synthesis of fatty
acids (7, 12-16).
One of these components has the rather unusual property of being relatively
stable to heat and dilute
hydrochloric
acid (7, 12, 13). Acetyl and malonyl moieties can
be transferred from their coenzyme A derivatives to this protein.
The acyl complexes are then condensed by suitable enzymes to
form acetoacetyl-enzyme,
which is subsequently
converted to
butyryl-enzyme
(15-19).
These findings support the scheme
proposed earlier for the synthetase from yeast (8) .I
Extracts from avocado mesocarp contain, in addition to the
previously studied particulate system (5,20), a soluble enzymatic
system, which converts malonyl coenzyme A to saturated long
chain fat’ty acids (21). This soluble enzyme has as one of its
components a heat-stable protein and therefore resembles the
bacterial systems. Some of the data discussed here have already
been reported elsewhere (22).
Plants
4104
Fat Metabolism
in Higher
Identi$cation
of Products of Fatty Acid Synthesis
Malonyl-CoA-CO2
Reaction
and
Fatty Acids-The
fatty acids were identified by a combination
of the following procedures.
The free fatty acids were separated
Fraction ARC, and
4 The terms Fraction IAV, Fraction IIAv,
enzyme IInc will be used to indicate the source of each fraction.
AV represents avocado enzymes, and EC, E. co&.
XXIII
Vol. 239, No.
12
by reverse phase chromatography
on siliconized Whatman No. 1
paper according to the method of Mangold, Lamp, and Schenk
(29) and Buchanan
(31) with the modification
of Yang and
Stumpf (21). Polar and nonpolar
acids were separated as
methyl esters by thin layer chromatography
on silicic acid with
n-hexane-diethyl
ether (10:6) as a solvent system (31). Saturated and unsaturated
acids were separated as methyl esters on
silicic acid thin layer plates impregnated with silver nitrate (32).
Gas-liquid
chromatography
was performed
with a Wilkens
Aerograph
A-90P instrument.
The conditions
are described
in the legend for Fig. 5. The effluent vapors were passed directly
into a Nuclear-Chicago
proportional
tube radioactive
counter
for continuous 14C monitoring.
Malonyl-CoA-CO2
Exchange-The
product of the exchange
reaction was identified as malonyl thioester by the following
criteria.
The samples of the experiments described in Table V
were saponified by heating for 10 minutes at 100” in 0.4 N KOH.
They were then acidified with HCl, and 100 mg of malonic acid
After an aliquot was counted on paper
were added as carrier.
in the scintillation
counter, the samples were passed through a
column (0.7 x 10 cm) of Dowex 50-H+ and lyophilized.
The
malonic acid was sublimed at 90” and a pressure of 0.05 mm of
Hg. The specific activity was determined, and the samples were
recrystallized
from ethyl acetate-petroleum
ether. The specific
activity was the same as before the recrystallization.
All the
radioactivity
fixed from 14C02 into nonvolatile
material was
associated with malonic acid.
After treatment
of the samples with hydroxylamine,
the
thioesters were converted to their hydroxamates
(26). After
removal of the salt by use of Dowex 50 in the way outlined above,
the samples were subjected to paper electrophoresis in pyridineacetic acid-water (100: 10 :890) for 2 hours at about 40 volts per
cm on Whatman No. 1 paper and to paper chromatography
in
pyridine-isopropyl
alcohol-water
(1: 1: 1) (33). In both systems
over 90 y0 of the radioactivity
moved to the same spot as authentic malonic monohydroxamate.
RESULTS
for Fatty Acid Synthesis-When
malonyl-COLA,
Requirements
acetyl-CoA, and a TPNH-generating
system are incubated with
varying amounts of a crude extract of avocado mesocarp, fatty
acid synthesis proceeds at an increasing rate. It was found that
DPNH meets the requirement of a dissociable cofactor, as shown
in Fig. 1. In the absence of DPNH an upward curvature is
obtained, whereas with the addition of DPNH the incorporation
increases linearly with increasing protein concentration.
After passage through a column of Sephadex G-25, the extract
showed several cofactor requirements
(Table I). The system is
completely dependent on TPNH.
Although
the requirement
for DPNH was not absolute in the crude ext’racts, it became
more pronounced after fractionation
of the system (see below).
Addition of boiled extract to the complete system resulted in a
65% stimulation.
A requirement
for acetyl-CoA
was not
demonstrable, presumably owing to the presence of malonyl-CoA
decarboxylase, which has been shown to occur in a variety of
plants, including avocado (34). Since it has already been established (21) that fatty acids are synthesized de novo, an elongation
mechanism with malonyl-CoA
is ruled out.
An enzyme dependence curve of the Sephadex-treated
extract
again showed an upward curvature, which could be overcome
by the addition of boiled crude extract.
Since all the other
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
The extract had 2 to 4 mg of protein per ml.
a Btichner funnel.
For the preparation
of boiled juice, the extract was heated for 3
minutes in a boiling water bath and separated from the precipitate by filtration.
In the experiment described in Table I, 20 ml of extract were
passed over a Sephadex G-25 column (medium grade, 2.5 x 25
M
cm, equilibrated
against 0.005 M potassium phosphate-O.001
2-mercaptoethanol,
pH 7.2); 25 ml of eluent with a protein
content of 2 to 3 mg per ml were collected.
In the experiments
described in Fig. 2 and Table II, 5.82 g of ammonium sulfate
were added with stirring to 20 ml of this solution (0 to 50%
saturation) in 30 minutes.
Throughout
this procedure the pH
was maintained
at 7 (adjustment
against Hydrion pH paper,
6.0 to 8.0, Micro Essential Laboratory)
by addition of ammonium hydroxide.
After 15 minutes the solution was centrifuged for 10 minutes at 37,000 X g. The precipitate
was
discarded.
To the clear supernatant
(22.4 ml), 4.4 g of ammonium sulfate were added under the conditions described above
The precipitate formed was centrifuged
(50 to 80% saturation).
for 10 minutes at 105,000 X g in a Spinco ultracentrifuge.
The
protein was dissolved in 4 ml of 0.02 M potassium phosphatewas used in Fig. 2
0.001 M cysteine, pH 7.3. This preparation
and Table II.
For the preparation
of avocado Fraction I and avocado Fraction II,4 15 kg of avocado mesocarp were worked up batchwise in
the following manner.
The crude extract was prepared from
I.5 kg of mesocarp as described above. The resulting 2.2 liters
of crude extract were saturated with ammonium
sulfate with
thorough stirring for 1 hour.
The pH was adjusted to 7 as
described above. After stirring for another hour, the precipitate
which formed was collected by centrifugation
for 30 minutes at
37,0C0 x g. The turbid supernatant was discarded.
The pellet
was dissolved in 80 ml of 0.1 M Tris-HCl buffer at pH 7.9, containThe buffer was changed after
ing 0.01 M 2-mercaptoethanol.
the first 4 hours. Ten such preparations
were pooled, yielding
1820 ml of concentrate which lost no activity when stored frozen for several months.
The above concentrate
(198 ml) was freed from insoluble
material by centrifugation
to give 178 ml of solution containing
18.7 mg of protein per ml. This solution was separated into
three fractions by addition of solid ammonium sulfate at pH 7.
The protein was sedimented by centrifugation
for 10 minutes
at 78,000 x g and taken up in a small amount of 0.05 M potassium phosphate-O.01
M 2-mercaptoethanol
at pH 7.2. The
fraction from 0 to 20% saturation was discarded, the fraction
from 20 to 609/; saturation was designated Fraction IAv, and
the fraction from 60 to 90% saturation was named Fraction
II*v.
Both fractions were dialyzed for 4 hours against 0.01 M
potassium phosphate-O.01
M 2-mercaptoethanol,
pH 7.9, and
then were stored frozen. Fraction IIAv was routinely heated
for 1 minute in a boiling water bath to destroy traces of Fraction
IAv activity.
Plants.
December
1964
P. Overath and P. K. Stumpf
4105
and was not dialyzable.
Table II summarizes some further
properties of the factor.
Whereas the cofactor is stable when
exposed to 100” at pH 6.9 with little loss in activity, it is destroyed when heated at pH 5.3. The activity was lost com-
I8OOC
+ BOILED
EXTRACT
-BOILED
MG
EXTRACT
4
3
I
*PROTEIN
1. Requirement
of DPNH
for fatty
acid synt.hesis.
The
complete
system
contained
potassium
phosphate
buffer,
pH 7.2,
100 rmoles;
malonyl-2-W-CoA
(83,000 c.p.m.),
0.104pmole;
acetylCoA, 0.1 pmole;
TPN+,
0.2 pmole;
glucose
6-phosphate,
1 pmole;
glucose 6-phosphate
dehydrogenase,
0.1 unit; and varying
amounts
of crude extract
from
avocado.
The two curves
were obtained
with different
extracts.
DPNH
(1 rmole)
was added as indicated;
total volume,
1.5 ml; time of incubation,
30 minutes
at 37”.
FIG.
TABLE
I
Cofactor requirement
Incubation
Fig. 1; enzyme
(see “Materials
conditions
and components
were
(2.9 mg) was used after treatment
and Methods”).
the
same
as in
MG
FIG. 2. Stimulation
of the (NH&S04
fraction
by boiled extract.
The complete
system
contained
potassium
phosphate
buffer,
pH
7.2, 40 pmoles; malonyl-2-‘4C-CoA
(29,500 c.p.m.), 0.038 pmole;
acetyl-CoA,
0.036 @mole; TPN+,
0.07 pmole;
glucose
6-phosphate,
0.3 rmole;
glucose
B-phosphate
dehydrogenase,
0.07 unit;
and
DPNH,
0.3 rmole.
The enzyme
was obtained
by (NH&SO4
fractionation
of the Sephadex-treated
crude
extract
(see “Materials
and Methods”).
Total volume,
0.5 ml; time of incubation,
30 minutes
at 37”.
with Sephadex
Properties
Incorporation
into
fatty acids
System
C.P.nZ.
Complete .....................................
-Acetyl-CoA.
.............................
-TPN+
- glucose-6-P
dehydrogenase.
- DPNH ....................................
-Enzyme
...................................
+Boiled
extract
(0.5 ml). ...................
cofactors
were
present
at saturating
8,424
9,630
376
4,148
46
13 ) 907
......
amounts,
another
dissociable
cofactor, which was present in the boiled extract, appeared to be
required.
In order to obtain an enzyme preparation which would
be completely dependent on boiled juice, the Sephadex-treated
extract was fractionated
with (NH4)zS04.
A fraction was obtained between 50 and 80% saturation which had essentially no
activity without addition of boiled extract.
In Fig. 2 the dependence of fatty acid synthesis on added boiled extract is shown.
With this (NHJzSOd fraction it was possible to study the
properties of the cofactor in the boiled extract in more detail.
The factor could not be replaced by flavin mononucleotide
or
flavin
adenine
dinucleotide.
The
activity
was
lost
on
ashing
PROTEIN
TABLE
II
of heat-stable cofactor
The complete
system
contained
the same components
as in
Fig. 1. The (NH4)&?04
fraction
(0.89 mg) (see “Materials
and
Methods”)
and boiled extract
(0.5 ml) were used.
Preincubation:
0.5 ml of boiled extract
with 0.15 mg of papain
in a total volume
of
0.55 ml for 30 minutes
at 37”.
system
Incorporation
into
fatty acids
C.p.W%.
Complete.....................................
+Boiled
extract
(3 min, loo”, at pH 6.9)
+Boiled
extract
(3 min, loo”, at pH 5.3).
+Dialyzed
boiled extract
(3 min, loo”, at pH
6.9)......................................
+Boiled
extract
(3 min, loo”, at pH 6.9) preincubated
with papain
and again heated
for
5 min at 100”.
. . .., ..,
+Boiled
extract
(3 min, loo”, at pH 6.9) preincubated
and again
heated
for 5 min at
loo”......................................
+Preincubated
with papain,
heated
for 5 min
at 100” + boiled extract
(3 min, loo”, at pH
6.9)......................................
134
1282
67
1596
4
877
1143
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
3000
Fat Metabolism
4106
TABLE
III
Separation
of two enzymatic
activities
fractionation
with (NH4)$304
in Higher
Plants.
XXIII
E. COLI
6000
by
Vol. 239, No.
ENZYME
12
ZiI = CONSTANT
Incubation conditions were the same as in Fig. 2, except that
the tubes contained 0.47 mg of Fraction I.&y and 0.11 mg of Fraction 11~~ where indicated.
Incorporation
into
fatty acids
Enzyme
/
C.$.rn.
161
18
2025
1243
6 One unit is defined as 1 mpmole of malonyl-CoA
fatty
acids
per minute
at 37” under
the conditions
FRACTION
AVOCADO
II= CONSTANT
I
19
pletely when the boiled extract was incubated with papain.
The
two last experiments in Table II show that this loss of activity
is due neither to the inactivation of the cofactor during the second
heat treatment nor to an inhibition
by the inactivated
papain.
It seemed likely, therefore, that a heat-stable protein similar to
the protein found in bacterial systems (7, 12) is involved in the
avocado synthetase.
Separation of Two Enxymatic Activities-The
procedure for
the preparation
of the (NHd)zSOd fraction used in Fig. 2 and
Table II proved to be rather difficult to reproduce.
The stimulation achieved by the addition of the boiled juice varied considerably. This was mainly caused by differences in the crude extracts,
which were prepared from several varieties of avocado, the exact
Furtherripening stage of which was difficult to determine.
more, accurate fractionations
were made impossible by the low
protein content of the crude extract (2 to 4 mg per ml).
These difficulties were overcome in the following way. The
crude extract was saturated with (NHI)tSO+
and the protein
was collected as quantitatively
as possible (49% yield) by centrifugation.
The protein was dissolved in a small volume of
Tris buffer and dialyzed (see “Materials
and Methods”).
In
this concentrated extract (19 mg of protein per ml), two fractions
could be separated by precipitation
with (NH,)aS04.
Table
III shows the result of a typical fractionation.
The heat-labile
fraction
(Fraction
IAv) previously
precipitated
from dilute
extracts between 50 and 80% saturation appeared now between
20 and 60% saturation.
The heat-stable
protein previously
lost by the (NHJ2S04 fractionation
was now precipitated
between 60 and 90% saturation (Fraction II*v).
After the separation of the two fractions, a determination
of
their activities in the crude extract was undertaken.
When an
excess of either Fraction IAv or IIAv was used under conditions
of fatty acid synthesis (see legend for Fig. 2 for details), it was
observed that in the crude extract, Fraction IAv (6.3 units5 per
ml; specific activity, 1.9), was about 8 times more active than
Fraction IIAv (0.78 unit per ml; specific activity, 0.23). However, the same assay is restrictive when used for the mutual
determination
of the two fractions themselves.
This is evident
from the curves with the solid dots in Figs. 3 and 4. In Fig. 3,
Fraction IAv was varied in the presence of an excess of Fraction
into
:
Ix
I
x I
I
incorporated
of Fig. 2.
I
1.0
I
0.5
MG
PROTEIN
FIG. 3. The dependence
of fatty
acid synthesis
on Fraction
IAV
in the presence
of an excess of Fraction
IIAV
or enzyme
&o.
The complete
system
was the same as in Fig. 2 with the exception
that 0.044 pmole
of malonyl-2-14C-CoA
(34,400 c.p.m.)
was used.
Enzyme
IIno (2.5 X 10F2 unit, 0.57 unit per mg of protein)
or Fraction 11.&v (1.1 mg of protein)
was added to the corresponding
tubes.
Fraction
IAy was added as shown on the abscissa.
z
P
”
.
x
20000
4000
,
t AVOCADO
FRACTION
I = CONSTANT
/’
4000
f
FIG. 4. The
in the presence
The complete
E. coli (1.7 X
Fraction
IAV
Fraction
IIAV
I
0.5
1
I.0
MG
1.5
PROTEIN
2.0
I
2.5
dependence
of fatty
acid synthesis
on Fraction
11~~
of an excess of Fraction
IAV
or Fraction
Asc.
system was the same as in Fig. 3. Fraction
A from
1OP unit, 10.4 X 10F3 unit per mg of protein)
or
(1.2 mg) was added
to the corresponding
tubes.
was added as indicated
on the abscissa.
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
Fraction IAJJ (20 to BOY0 saturation).
.
Fraction 11~~ (60 to 90% saturation).
Fraction IAv + Fraction IIAv. _.
Fraction Inv f boiled Fraction &V (3 min,
1000)...........
Boiled Fraction I*v + Fraction 11~~ (3 min,
loo”).........................................
:
December
1964
P. Overath and P. K. Xtumpf
TABLE
Mutual
replacement
IV
fronr
avocado
and E. coli
for fatty acid synthesis
The conditions
were the same as in Fig. 2 except
that
the
amount
of malonyl-2-K-CoA
was 0.061 pmole
(38,000
c.p.m.)
and
5 pmoles
of 2-mercaptoethanol
were
added.
Enzymes:
Fraction
AEC, 1.7 X 10W3 unit
(10.4 units
per mg of protein);
Enzyme
IIEc, 1.3 X lo+ unit (0.57 unit per mg of protein);
Fraction IAV, 1 mg; and Fraction
II.ty,
0.6 mg.
The samples
were
incubated
for 60 minutes
at 37”.
of fractions
Enzyme
Product
c.p.m.
Fraction
Enzyme
Fraction
Fraction
Fraction
Fraction
Fraction
Fraction
Fraction
Enzyme
Azo
IIsc
IAV
11~~
Asc +
1~”
Asc
Azc
IAV
11~~
+
+
+
+
+
164
2
86
enzyme
IIsc
Fraction
Fraction
Fraction
enzyme
Fraction
II*1
IAV
II*1
I&c
II~T
26,69:
‘&:I*
(49%‘0), CIC:O (20%),
cl*:0 (20%)
fi.“.@a%),
c1s:o (18%)
1,303
3c
2,20
4,62-
- .“.”
(90%)
(27%)
O/
* The subscript
preceding
the colon denotes
bon atoms;
the subscript
following
the colon,
ble bonds.
et al. (13), it was found
. compounds
(73%))
c18:o
the number
the number
of carof dou-
that the main product of the E. coli
synthetase is a monounsaturated
fatty acid with 18 carbon atoms,
presumably
vaccenic acid. In addition, stearic and palmitic
acids were formed in about equal amounts.
All three chromatographic methods mentioned above lead to the same conclusion.
Fraction IAV + Fraction IIAA-In
agreement with Yang and
Stumpf (21), it was found that the products of the avocado
synthetase are stearic and palmitic acids. It might be mentioned, however, that the ratio Cig:& is about 4 in the crude
extract, whereas it is about 0.5 under the conditions used in
Table IV.
Fraction
ABC + Fraction
IIAa-It
has been shown (15, 16,
18, 19) that enzyme I& functions as the carrier for the intermediates whereas Fraction j 4 no supplies all the heat-labile enIf one
zymes necessary for fatty acid synthesis in E. coli.
assumes the same to be true for the avocado fractions, then a
combination
of Fraction AEC and Fraction IIAv should yield
vaccenic, stearic, and palmitic acids. Surprisingly,
the expected
results were not obtained.
In reverse phase paper chromatography,
the product (free
acids) moved with an RF of 0.89 in a solvent system of 85%
acetic acid, indicating that the compound (or compounds) was
polar in nature.
This was confirmed by thin layer chromatography of the methyl esters on silica gel G with hexane-diethyl
ether
(10:6) as a solvent; most of the radioactivity
had an RF of 0.26.
This RF was identical with authentic P-hydroxy acid methyl
esters with 10 to 14 carbon atoms. When this spot was eluted
and subjected to thin layer chromatography
on AgNOB, part
of the polar methyl esters were retained near the origin.
This
would indicate that a proportion
of the polar acids were unFig. 5 shows gas-liquid chromatograms of the methyl
saturated.
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
ILV. The first part of this curve shows an upward curvature,
which makes an accurate determination
of Fraction IAv impossible.
The shape of the curve would indicate another dissociable cofactor.
Additions
of flavin mononucleotide,
boiled
extract, 2-mercaptoethanol,
or bovine serum albumin were tested,
but proved to be without effect. In Fig. 4, Fraction IIAv was
varied in the presence of an excess of Fraction IAv. Whereas
the first part of the curve seems to be fairly linear, it falls off
rather abruptly with higher concentrations
of Fraction IIAv.
This would suggest an inhibitor in Fraction II*v, which probably
also decreases the rate in the first part of the curve (see below).
Thus Fraction IAv can be assayed with Fraction IIEc but not
with Fraction II*v, and Fraction IIAv can be assayed with
Fraction IEC but not with Fraction IAv. Interactions appear to
occur when Fractions IAkv and IIAv are combined at high concentrations.
The nature of these interactions awaits the further
purifications of the two fractions.
Mutual Replacement of Fractions from Avocado and E. coli in
Fatty Acid Synthesis-As
mentioned previously, a heat-stable
(enzyme II&
and a heat-labile (Fraction Ano) protein fraction
were shown to be required for fatty acid synthesis in E. coli (13).
From the comparative point of view, it seemed of considerable
interest to study the cross-reactions of the bacterial enzymes
with the avocado fractions.
Moreover, the E. coli fractions
would be useful for the purification
of the avocado enzymes.
Table IV summarizes an experiment in which the two fractions
from avocado were combined with Fraction AEC and enzyme
It is seen that all fractions alone and the combinations
&c.
Fraction AEC + Fraction IAv and enzyme IIno + Fraction
II *v give little or no activity which is extractable with chloroform. The combinations of Fraction AEc + Fraction IIAv and
Fraction IAv + enzyme IIEc give good activity.
This result
suggests, therefore, that the two heat-stable proteins have the
same function.
The curves marked by crosses in Figs. 3 and 4 show the enzyme
dependences of Fraction IAv and Fraction IIAv in the presence
of an excess of enzyme IIno and Fraction Ano, respectively.
The shape of the curve in Fig. 3 seems to be the same as the
Fraction IAv, however,
corresponding curve with Fraction II*“.
is somewhat more active in the presence of enzyme I&.
The
leveling off of the curve with enzyme I& is not related to a
saturation effect. This becomes evident when the same amount
of enzyme II,o is tested with Fraction Ano. As shown already
by Goldman et al. (13), the fatty acid synthesis of E. coli is
linearly dependent on Fraction AEC in the presence of an excess
of enzyme I&.
Under the conditions of Fig. 3, such a dependence levels off after an incorporation
3 times greater than
that catalyzed by Fraction IAv. One would thus conclude that
Fraction IAv becomes inhibitory at higher concentrations.
In the presence of an excess of Fraction AEC (Fig. 4, curve
marked with crosses), a linear relationship
exists between the
concentration of Fraction IIAv and incorporation
of malonyl-CoA
over a considerable range. The activity of Fraction IIAv is
greater in the presence of Fraction AEC than in the presence of
Fraction IAv.
Analysis of Products (see Table IV)-The
products of the
various combinations
shown in Table IV were analyzed for the
type and the amount of fatty acids formed.
A combination
of
thin layer, gas-liquid, and paper chromatography
was used.
Fraction IEC + Enzyme II Bc-In agreement with Goldman
4107
4108
Fat Metabolism
in Higher
-L
INJECTION
INJECTION
I
GLC
Vol. 239, No.
XXIII
Malonyl-CoA-CO2
Exchange
Reaction-The
acyl-CoA-dependent exchange of malonyl-CoA
and CO2 proved to be a very
useful tool for the elucidation of the mechanism of the condensation reaction (12, 13). In analogy to the behavior of the enzymes
from E. coli and C. kluyveri, it could be expected that the malonylCoA-CO2 exchange reaction would also be operative with the
avocado synthetase and that moreover both fractions would be
required.
This reaction would also provide a simple and rapid
assay for the purification
of the heat-stable
Fraction
II*v.
Table V summarizes the results of attempts to demonstrate the
exchange reaction in the avocado system and cross-reactions
with the corresponding
E. coli fractions.
In the presence of
phosphate buffer at pH 6.3, the various fractions alone show
little or no activity.
Combination
of the two fractions from
avocado gives no activity.
With the exception of the combined
E. COG fractions, the only system which gives exchange is the
combination
of Fraction Ano with Fraction II*v.
The reaction
Experiments in
is dependent on malonyl-CoA
and caproyl-CoA.
imidazole
buffer give essentially the same general picture as
with phosphate buffer.
However, now Fraction IAv catalyzes
an exchange activity, which is not dependent on the presence of
either Fraction IIAv or enzyme 11,o. It is also worth mentioning that since Fraction IAv has little if any inhibitory effect when
added to the exchange system of the combined E. coli fractions,
the lack of exchange activity of the type described by Alberts
TABLE
Mutual
(ZOLUMN:
DEGS
FIG.
5. Gas-liquid
chromatography
of the methylated
compounds
formed
by the combination
of Fraction
AEC and Fraction
11~~.
The carrier
Peaks I, II, and III are the methyl
esters of
@-hydroxydecanoic,
P-hydroxylauric,
and
fi-hydroxymyristic
acids, respectively.
The SE-30 column
(5 feet, 2 inch,
stainless
steel)
was operated
at 222”.
The diethylene
glycol
succinate
(DEGS)
column
(6 feet, $ inch, stainless
steel) had a temperature
of 182”.
Effluents
from an Aerograph
A-90P were passed into a
Nuclear-Chicago
continuous
radioactive
monitoring
unit,
and
both the gas-liquid
chromatographic
(GLC)
mass peaks and the
radioactive
peaks were recorded
by a Texas Servo-Rite
dual chart
recorder.
ester of the product
on two columns
with concomitant
i4C tracings
by a radioactive
monitor
system.
On the SE-30
column,
most
of the radioactivity
had the same retention
times as the carrier
peaks P-hydroxylauric
acid methyl
ester
(Peak
II) and P-hyOn the diethylene
droxymyristic
acid methyl
ester (Peak III).
of these peaks was
glycol succinate column, a further resolution
obtained.
Smaller
peaks were now observed
near Peaks II and
III.
The slower moving
peaks behind
II and III again would
indicate
unsaturation.
P-Hydroxymyristic
acid and /3-hydroxylauric
acid6 have been definitely
identified
by recrystallization
carrier acids to constant specific activity (solvents,
with authentic
ethanol-Hz0
and hexane).
cation
other
of the
Fraction
I,,
+
Further
work will lead to the identifi-
peaks.
Enzyme
II Bc-In
this combination,
the en-
zymes in Fraction
IAV would
be expected
to determine
the type
of fatty
acids synthesized.
It was observed
that the products
were almost identical with the product of
of the combination
the combination
of both avocado
fractions,
namely,
palmitic
acid.
6 We are greatly
indebted
ance in these experiments.
to Mr.
Robert
Simoni
for
his assist-
12
V
from avocado and E. coli
replacement
of fractions
for malonyl-CoA-CO2
exchange
reaction
The complete
system
contained
50 Mmoles
of potassium
phosphate
buffer
at pH 6.3 or 50 pmoles
of imidazole-HCl
buffer
at
pH 6.2.
The other components
were the same as described
with
by Alberts
et al. (12) the exception
that only 5 pmoles
of 2.mercaptoethanol
were used in the experiment
with phosphate
buffer.
Enzymes:
Fraction
Ant, 0.9 X 10m3 unit (4.3 X low3 unit per mg of
protein);
enzyme
IInc, 2.6 X 10e3 unit (74 X 10e3 unit per mg of
protein);
Fraction
I*v, 0.5 mg; and Fraction
IIav,
0.6 mg.
The
samples
were incubated
for 60 minutes
at 30” and counted
as described
in “Experimental
Procedure.”
Malonyl-CoA-COz
reaction
Fraction
Enzyme
Fraction
Fraction
Fraction
Fraction
Fraction
Fraction
Fraction
1
AeC.
11~~.
IAN,.........................
1I.~v.........................
AEC + enzyme
1~v + Fraction
Aec + Fraction
AEC + Fraction
I*v + enzyme
Enzyme I1~c + Fraction
Fraction
AEC + enzyme
nyl-CoA............................
Fraction
Ant + enzyme
royl-CoA..
Fraction
Ant + Fraction
nyl-CoA.
Fraction
AEC + Fraction
royl-CoA.
11~~
1I~v.
1~v..
11~~.
11~~.
11~~.
11~~ -
Phosphate
buffer, pH 6.3
Imidazole
buffer, pH 6.2
c.p.m.
c.p.m.
246
0
276
0
3900
291
629
4610
368
0
410
0
2130
0
8250
2110
2640
4850
2340
28
malo0
IInc
-
11~”
- malo
1I~v
-
exchange
in
cap1140
0
cap964
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
GLC
Plants.
December
1964
et al. (12) cannot be ascribed to possible inhibition
by avocado
proteins.
The product of the reaction of Fraction IAv alone and the
combination
of Fraction Ano and Fraction IIAv (incubation
in
imidazole buffer under conditions identical with those in Table
V) is a malonyl thioester, presumably malonyl-CoA.
Saponification of the product with alkali and subsequent isolation of the
labeled malonic acid in the presence of carrier showed that
essentially all the activity remained in this acid. After conversion of the product to the hydroxamate,
over 90% of the
radioactivity
behaved like malonic monohydroxamate,
as shown
by paper chromatography
and paper electrophoresis.
The demonstration
of an exchange reaction with Fraction Ano
and Fraction IIAv provides a simple assay for the determination
6000
7000
6000
.
t
2
4
5000
:
z
4000
2
::
w
f
t?
x
c
50”
CJ
3000
2000
IOOC
I
I
1
I
0.2
0.4
0.6
0.6
MG
PROTEIN
I
1.0
I
I
1.2
t
1.4
1.6
6. Dependence of the malonyl-CoA-CO*
exchange reaction
on Fraction 11~~ in the presence of an excess of Fraction AEC.
The incubation conditions were the same as described by Alberts
et al. (12); 3.4 X 1OW unit of Fraction AEC (10.4 X IO-3 unit per
mg of protein) and the indicated amounts of Fraction IIAv were
added.
FIG.
TABLE
Replacement
of CoA
by pantetheine
VI
derivatives
in exchange
reaction
The complete system contained the same components as given
by Alberts et al. (12); 0.7 rmole of malonyl thioester and 0.1
pmole of caproyl thioesters were added. Enzymes: Fraction
AEC, 2.0 X 10W3unit (10.4 X 1OW unit per mg of protein); Fraction
II*“, 0.5 mg of protein.
Radioactivity
in
malonyl
thioester
c.p.m.
Malonyl-CoA
+ caproyl-CoA.
.
Malonylpantetheine
+ caproylpantetheine.
Malonylpantetheine.
Caproylpantetheine
3140
5400
330
220
of Fraction IIAv.
It can be seen from Fig. 6 that the exchange is
linearly dependent on Fraction IIAv in the presence of an excess
of Fraction Ano. As in the bacterial systems (34), the Coh
derivatives can be replaced by the corresponding
pantetheine
Table VI summarizes these data.
derivatives.
DISCUSSION
Unlike the animal and yeast fatty acid synthetases, but like
the bacterial systems, the soluble synthetase from avocado
mesocarp is not a single complex but can be resolved into at
The heat-stable Fraction II of the plant
least two components.
system has stability properties strikingly
similar to enzyme
I&
(13). In both the malonyl-CoA-CO2
exchange reaction
and the fatty acid synthetase system, Fraction IIAv replaces
enzyme I& in the system in which enzyme AEC serves as the
source of heat-labile enzymes.
Conversely, enzyme IIEG fully
substitutes for Fraction IIAv in a fatty acid-synthesizing
system
with Fraction IAV as the source of the heat-labile enzymes.
However, in the fatty acid-synthesizing
system consisting
of enzyme AEC and Fraction IIAv, the final products are a mixture of P-hydroxy long chain fatty acids rather than the expected
products, namely, vaccenic, stearic, and palmitic acids. This
unexpected result suggests that Fraction IIAv inhibits an enzyme
in enzyme AEc which normally converts the ,&hydroxy
acids
to long chain fatty acids. Likely candidates for such an inhibition are enzymes II&
or IVEc derived from Fract.ion AEc, the
absence of which gives rise to polar acids (16).
Although both fractions from avocado are required for fatty
acid synthesis, the requirement of both fractions for the malonylCoA-COa-exchange
could not be demonstrated.
The reason
for this remains obscure.
The condensation of ace@CoA
and
malonyl-CoA
must occur under the conditions for fatty acid
synthesis.
Since Fraction IIAv catalyzes the exchange effectively when combined with Fraction AEC, Fraction IAV may be
The kinet,ics of Fig. 3 suggests a strong
the site of the difficulty.
inhibitory
effect by Fraction IAV. At least two possibilities
may exist: (a) the /3-ketoacyl intermediate
is inactivated,
and
in the presence of TPNH the reduction might be faster than the
inactivation,
or (b) the condensing enzyme in Fraction IAV may
not catalyze the back reaction at all or at the required rate.
Since the malonyl-CoA-COz
exchange reaction with Fraction
Ano depends on the presence of Fraction IIAv, a rapid assay for
the purification
of the heat stable enzyme from avocado is now
available.
Furture investigations
will be directed toward the
comparative aspects of the plant and the bacterial synthetases.
It is of interest that lettuce chloroplast preparations incorporate
malonyl-CoA
to a slight extent in the absence of the heat-stable
protein, but when either Fraction IIAv or enzyme II,o is added,
marked incorporation
is noted.7 The presence of the heat-stable
protein would seem to be a limiting factor in the capacity of
chloroplasts to synthesize fatty acid.
SUMMARY
1. The soluble fatty acid synthetase from avocado mesocarp
has been fractionated
into a heat-labile
(Fraction IAV) and a
heat-stable
(Fraction
IIAv)
component.
Either
component
alone is inactive in condensing malonyl coenzyme to form long
chain fatty acid, but when combined they catalyze the synthesis
of palmitic and stearic acids.
7 J. Brooks and P. K. Stumpf, unpublished
observations.
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
9000
z
s
4109
P. Overath and P. K. Stumpf
Fat Metabolism in Higher Plants.
4110
2. Fraction IIAv appears to be a heat-stable protein which
can replace enzyme I&o in the Escherichia coli synthetase system
and the malonyl coenzyme A-CO2 exchange.
3. When Fraction IIAv is added to Fraction Ano, P-hydroxylauric and P-hydroxymyristic
acids are the major products
rather than the expected products for Fraction Ano, namely,
vaccenic, palmitic, and stearic acids.
REFERENCES
1. BRADY,
2. WAKIL,
3.
4.
6.
7.
8.
9.
10.
Chem.,
235, 3093
(1960).
11. BRESSLER,
R., AND WAKIL,
S. J., J. Biol. Chem., 236, 1643
(1961).
12. ALBERTS, A. W., GOLDMAN,
P., AND VAGELOS, P. R., J. BioZ.
Chem., 238, 557 (1963).
13. GOLDMAN,
P., ALBERTS, A. W., AND VAGELOS, P. R., J. BioZ.
Chem.,
238, 1255 (1963).
Vol. 239, No.
12
14. NORRIS, A. T., AND BLOCH, K., J. BioZ. Chem., 238, PC3133
(1963).
15. ALBERTS, A. W., Federation
Proc., 23, 166 (1964).
16. PUGH, E. L., SAUER, F., AND WAKIL, S. J., Federation
Proc.,
23, 166 (1964).
P., AND VAGELOS,
P. R., Biochem.
and Biophys.
17. GOLDMAN,
Research
Communs., 7, 414 (1962).
18. GOLDMAN,
P., ALBERGS, A. W’., A&D VAGELOS, P. R., J. BioZ.
Chem., 238, 3579 (1963).
19. MAJERUS,
P. W., AND VAGELOS, P. R., Federation
Proc.,
23,
166 (1964).
20. STUMPF, P. K., AND BARBER, G. A., J. BioZ. Chem., 227, 407
(1957).
21. YANG, S. F., AND STUMPF, P. K., Biochim.
et Biophys.
Acta,
84, No. 6 (1964).
22. OVERATH.
P.. AND STUMPF, P. K., Federation
Proc..
23, 166
(1964). ’
23. EGGERER, H., AND LYNEN, F., Biochem. Z., 335,540 (1962).
24. ANTENRIETH.
W.. Ber. deut them. Ges.. 34, 182 (1901).
25. SIMON, E. J.‘, AND SHEMIN, D., J. Am. Chem.‘Soc:,
75, 2520
(1953).
26. STADTMAN,
E. R., in S. P. COLOWICK
AND N. 0. KAPLAN
(Editors), Methods in enzymology,
Vol. III, Academic Press,
Inc., New York, 1957, p. 938.
27. BEISENHERZ,
G., BOLTZE,
H. J., BUCHER,
T., CZOK,
R.,
GARBADE,
K. H., MEYER-ARENDT,
E., AND PFLEIDERER,
G., 2. Naturforsch.,
Pt. b, 8, 555 (1953).
28. KALCKAR, H. M., J. Biol. Chem., 167, 461 (1947).
H. K., LAMP, B. G., AND SCHENK, H., J. Am. Chem.
29. MANGOLD,
Xoc., 77, 6070 (1955).
30. BUCHANAN,
M. A., Anal. Chem., 31, 1616 (1959).
31. MANGOLD,
H. K., J. Am. Oil Chemists’
Sot., 38, 708 (1961).
32. MORRIS, L., Chem. and Ind. (London),
1238 (1962).
33. VAGELOS, P. R., J. Biol. Chem., 235, 346 (1960).
37, 121
34. HATCH, M. D., AND STUMPF, P. K., Plant Physiol.,
(1962).
Downloaded from www.jbc.org at UNIV OF CALIFORNIA RIVERS, on February 9, 2013
5.
R. O., Proc. N&Z. Acad. Xci. U. S., 44,993 (1958).
S. J., AND GANGULY,
J., J. Am. Chem. Sot., 81, 2597
(1958).
LYNEN, F., J. Cellular
Comp. Physiol.,
54 (Suppl. I), 33 (1959).
MARTIN,
D. B., HORNING,
M. G., AND VAGELOS, P. R., J. BioZ.
Chem., 236, 663 (1961).
BARRON.E.
J.. SQUIRES. C.. AND STUMPF. P. K.. J. BioZ. Chem..
236, 2810 (196lj.
’ ’
GOLDMAN,
P., ALBERTS, A. W., AND VAGELOS, P. R., Biochem.
and Biophys.
Research
Communs.,
5, 280 (1961).
LENNARZ,
W. J., LIGHT. R. J., AND BLOCH, K., Proc. Natl.
Acad. L&i. U. k., 48, 840 (1962).
LYNEN.
F.. Federation
Proc.. 20. 941 (1961).
LYNEN;
F.; in S. P. COLOW&K
END 6. O.‘KAPLAN
(Editors),
Methods
in enzymology,
Vol. V, Academic Press, Inc., New
York, 1962, p. 443.
BRADY,
R. O., BRADLEY,
R. M., AND TRAMS, E. G., J. Biol.
XXIII